Secondary estimation method for an atmospheric transmissivity image in a defogging process

Abstract

The invention discloses a secondary estimation method for an atmospheric transmissivity image in a defogging process. The secondary estimation method comprises the following steps: acquiring an atomization degraded image in haze weather; Obtaining a dark channel image by using region minimum filtering; Obtaining a global atmospheric light value and the original scene transmissivity of the image according to the dark channel image; Filtering the original scene transmissivity by adopting a guide filter to obtain coarse estimation of the scene transmissivity; Performing minimum variance median guided filtering on the rough estimation of the scene transmissivity to obtain the final scene transmissivity; And restoring the atomized degraded image by using the final scene transmissivity. Scene transmissivity is secondarily estimated through minimum variance median guided filtering, internal texture information of the image can be effectively reserved, the problems that texture information andedge information of the restored image are lost when a traditional dark channel defogging algorithm is used for obtaining the restored image are solved, the image restoration quality is improved, andcompared with a traditional algorithm, the internal information of the image is better reserved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for secondary estimation of atmospheric transmittance images in the process of defogging. Background technique [0002] With the rapid development of digital image acquisition technology and processing technology, images have become important information for people to obtain information. Images contain a lot of texture information and edge information, such as target tracking and detection in machine vision, image stitching technology, etc. The acquisition of information in the image is required. [0003] However, due to the impact of air quality in recent years, frequent occurrence of haze weather, haze weather reduces visibility, has a great impact on people's daily life and travel, and also reduces the quality of images collected by some equipment, thus for The acquisition of texture, edge and other information has a great impact. Moreover, image degradatio...

AI Technical Summary

Problems solved by technology

[0004] In the traditional dark channel dehazing algorithm, after the atmospheric transmittance matrix is ​​obtained, the original image is used as a template to perform a guided filter to obtain the corrected atmospheric transmittance matrix. However, since the guided filter is based on the template image and the image to be filtered The difference between the images to be filtered is used as the filtering standard, ignoring the image structure relationship between the template image and the image to be filtered

Images